| A | B |
|---|
| Heteroskedasticity | When error variances differ across observations. |
| Autocorrelation | When error covariances are not zero across observations. |
| Unbiasedness | On average estimator equals actual value. |
| Gauss-Markov Theorem | Under the Full Ideal Conditions, OLS estimates are Best Linear Unbiased Estimators. |
| t-test | Test of statistical significance for a single estimate. |
| F-test | Statistical test useful for joint hypotheses. |
| Source of bias, inconsistency and inefficiency | Omitted Variables Problem |
| Source of loss in efficiency | Autocorrelation/Heteroskedasticity |
| Type II error | Falsely failing to reject the null hypothesis |
| Type I error | Falsely rejecting the null hypothesis. |
| Null Hypothesis | Prior disbelief designed to be contradicted. |
| White Test | General test for heteroskedasticity when source is unknown. |
| Goldfeld-Quandt Test | Test for heteroskedasticity that entails breaking the data into subsamples. |
| Variance | Measure of linear dispersion around the mean. |
| Covariance | Measure of linear association of two random variables. |
| Dummy Variable | A variable that takes a value of zero or one. |
